🔌 Plug-and-Play RAG Framework — GoLang

A plug-and-play Retrieval-Augmented Generation (RAG) framework built in Go, enabling seamless switching between vector databases and LLMs, with support for text and PDF-based context ingestion. Designed for developers and researchers building intelligent systems without handling backend orchestration manually.

---

🛠 Tech Stack

• Language: Go (Golang)
• Vector Databases: Qdrant (REST API), Weaviate (GraphQL)
• LLMs: OpenAI GPT-4, Mistral 7B (API-based)
• Embeddings: External Hugging Face model (via ghcr.io)
• Tools: Docker (containerization), CLI-based interface

---

⚙️ Features

• 🔄 Modular CLI system with plug-and-play components
• 📄 Ingest context via raw text or PDF files
• 🧠 Choose between Qdrant and Weaviate for vector storage
• 🗣 Use OpenAI or Mistral as the LLM engine
• 🔐 Secure API key integration for LLM access
• 📦 Embeddings generated via external Hugging Face model pulled from ghcr.io

---

🚀 Usage Flow

1. Provide Context

→ Enter plain text or upload a PDF

2. Choose Vector DB

→ Select between Qdrant or Weaviate

3. Choose LLM

→ Select between OpenAI or Mistral (provide API key)

4. Ask Your Query

→ System retrieves relevant context and generates a response using the selected LLM

---

📈 Roadmap

• 🔧 Transition from CLI tool to a full-fledged package
• 🌐 Integrate with Gofr framework (open-source collaboration) for routing, logging & config management
• 🧪 Add unit tests, CI/CD, and packaging for public release